Monthly collection

August 2025

The papers that drew the most attention during this month of AI research.

Papers: 63
Citations: 6.5K
Research labs: 3

Explore the year

View all months →

Aug2025 Sep2025 Oct2025 Nov2025 Dec2025 Jan2026 Feb2026 Mar2026 Apr2026 May2026 Jun2026 Jul2026 Aug2026

1,016papers
curated and indexed

63 papers published in August 2025

01

Annual Meeting of the Association for Computational Linguistics

A.S.E: A Repository-Level Benchmark for Evaluating Security in AI-Generated Code

The paper introduces A.S.E (AI Code Generation Security Evaluation), a repository-level benchmark for assessing the security of AI-generated code. It is built from 120 instances derived from 40 real-world GitHub repositories with documented CVEs, expanded via semantic and structural mutations. The benchmark covers four vulnerability types (SQL injection,…

Keke Lian, Bin Wang, Lei Zhang, Libo Chen, et al.

Published: Aug 2025
Upvotes: 350
Citations: 15

Read summary Original

02

Meta AI

DINOv3

DINOv3 is a self-supervised vision foundation model that scales to 7B parameters, trained on a curated dataset of 1.689 billion images (LVD-1689M) from Instagram, combined with ImageNet-1k. The model uses a ViT architecture with RoPE embeddings and constant hyperparameter schedules. A key contribution is Gram anchoring, a regularization technique that…

Oriane Siméoni, Huy V. Vo, Maximilian Seitzer, Federico Baldassarre, et al.

Published: Aug 2025
Upvotes: 315
Citations: 1.2K

Read summary Original

03

Qwen

Qwen-Image Technical Report

Qwen-Image is a new image generation foundation model from the Qwen team, designed to excel in complex text rendering and precise image editing. It uses a comprehensive data pipeline and a progressive curriculum learning strategy to enhance text rendering, especially for Chinese. For editing, it employs a multi-task training paradigm with dual-encoding…

Chenfei Wu, Jiahao Li, Jingren Zhou, Junyang Lin, et al.

Published: Aug 2025
Upvotes: 277
Citations: 875

Read summary Original

04

Research paper

Intern-S1: A Scientific Multimodal Foundation Model

Intern-S1 is a multimodal Mixture-of-Experts (MoE) foundation model with 28 billion activated parameters (241B total), designed to bridge the gap between open-source and closed-source models in scientific domains. It was continually pre-trained on 5T tokens, including over 2.5T scientific tokens, using pipelines for PDF parsing, domain-centric web data,…

Lei Bai, Zhongrui Cai, Yuhang Cao, Maosong Cao, et al.

Published: Aug 2025
Upvotes: 274
Citations: 13

Read summary Original

05

arXiv.org

InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency

InternVL3.5 is a new family of open-source multimodal models that improves versatility, reasoning, and efficiency over its predecessor InternVL3. Key innovations include Cascade Reinforcement Learning (Cascade RL), which combines offline RL (MPO) for stable convergence and online RL (GSPO) for refined alignment, boosting reasoning performance by up to…

Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, et al.

Published: Aug 2025
Upvotes: 225
Citations: 1.2K

Read summary Original

06

Z.ai / GLM

GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models

GLM-4.5 is an open-source Mixture-of-Experts (MoE) large language model with 355B total and 32B activated parameters, developed by Zhipu AI and Tsinghua University. It features a hybrid reasoning method supporting both thinking and direct response modes. Trained on 23T tokens with multi-stage pre-training and mid-training (including repo-level code,…

GLM-4. 5 Team, :, Aohan Zeng, Xin Lv, et al.

Published: Aug 2025
Upvotes: 213
Citations: 409

Read summary Original

07

arXiv.org

On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification

This paper analyzes the generalization gap between Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) for Large Language Models. The authors show that the SFT gradient can be interpreted as a policy gradient with an implicit, sparse reward that is inversely proportional to the model's token probability, leading to unstable updates and poor…

Yongliang Wu, Yizhou Zhou, Zhou Ziheng, Yingzhe Peng, et al.

Published: Aug 2025
Upvotes: 190
Citations: 112

Read summary Original

08

arXiv.org

VibeVoice Technical Report

VIBEVOICE is a novel model from Microsoft Research for synthesizing long-form, multi-speaker speech (up to 90 minutes, 4 speakers) using next-token diffusion. It introduces a causal speech tokenizer with a 3200x compression rate (7.5 Hz frame rate), which is 80 times more efficient than Encodec while maintaining comparable performance. The model uses a…

Zhiliang Peng, Jianwei Yu, Wenhui Wang, Yaoyao Chang, et al.

Published: Aug 2025
Upvotes: 177
Citations: 36

Read summary Original

09

Research paper

VeriWeb: Verifiable Long-Chain Web Benchmark for Agentic Information-Seeking

VeriWeb is a new benchmark for evaluating web agents on long-horizon, information-rich tasks. It addresses limitations of prior benchmarks that focus on single-fact retrieval and outcome-only verification. VeriWeb emphasizes long-chain complexity (breadth- and depth-oriented search) and subtask-level verifiability, decomposing tasks into interdependent…

Shunyu Liu, Minghao Liu, Huichi Zhou, Zhenyu Cui, et al.

Published: Aug 2025
Upvotes: 164
Citations: 9

Read summary Original

10

arXiv.org

Memento: Fine-tuning LLM Agents without Fine-tuning LLMs

The paper introduces Memento, a learning paradigm for LLM agents that enables continual adaptation without fine-tuning the underlying LLM. It formalizes a Memory-augmented Markov Decision Process (M-MDP) with a neural case-selection policy, storing past experiences in an episodic case bank. The policy is updated via online soft Q-learning, with retrieval…

Huichi Zhou, Yihang Chen, Siyuan Guo, Xue Yan, et al.

Published: Aug 2025
Upvotes: 162
Citations: 87

Read summary Original

11

arXiv.org

We-Math 2.0: A Versatile MathBook System for Incentivizing Visual Mathematical Reasoning

WE-MATH 2.0 is a unified system to enhance multimodal large language models' (MLLMs) mathematical reasoning. It integrates a structured knowledge system, model-centric data modeling, and reinforcement learning (RL). The MathBook Knowledge System is a five-level hierarchy with 491 knowledge points and 1,819 fundamental principles. MathBook-Standard is a…

Runqi Qiao, Qiuna Tan, Peiqing Yang, Yanzi Wang, et al.

Published: Aug 2025
Upvotes: 146
Citations: 30

Read summary Original

12

arXiv.org

NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at Scale

NextStep-1 is a 14B autoregressive model for text-to-image generation that uses continuous image tokens instead of discrete VQ tokens, paired with a 157M flow matching head. It is initialized from Qwen2.5-14B and trained with next-token prediction, combining cross-entropy loss for text and flow matching loss for images. The image tokenizer, fine-tuned from…

NextStep Team, Chunrui Han, Guopeng Li, Jingwei Wu, et al.

Published: Aug 2025
Upvotes: 146
Citations: 52

Read summary Original

13

Qwen

WebWatcher: Breaking New Frontier of Vision-Language Deep Research Agent

WebWatcher is a multimodal deep research agent introduced by Alibaba's Tongyi Lab to address the limitations of text-centric web agents by integrating visual information and complex reasoning. It uses high-quality synthetic multimodal trajectories for cold-start training, multiple tools (web search, image search, webpage visit, code interpreter, OCR), and…

Xinyu Geng, Peng Xia, Zhen Zhang, Xinyu Wang, et al.

Published: Aug 2025
Upvotes: 144
Citations: 101

Read summary Original

14

arXiv.org

A Survey of Scientific Large Language Models: From Data Foundations to Agent Frontiers

This survey reframes the development of Scientific Large Language Models (Sci-LLMs) as a co-evolution between models and their data substrate, providing a data-centric synthesis across six scientific domains (physics, chemistry, materials science, life sciences, astronomy, and Earth science). It introduces a unified taxonomy of scientific data and a…

Ming Hu, Chenglong Ma, Wei Li, Wanghan Xu, et al.

Published: Aug 2025
Upvotes: 144
Citations: 26

Read summary Original

15

arXiv.org

Agent Lightning: Train ANY AI Agents with Reinforcement Learning

Agent Lightning is a framework that enables reinforcement learning (RL) training of large language models (LLMs) for any AI agent, achieving complete decoupling between agent execution and training with almost zero code modifications. It formulates agent execution as a Markov decision process (MDP), defining a unified data interface that abstracts agent…

Xufang Luo, Yuge Zhang, Zhiyuan He, Zilong Wang, et al.

Published: Aug 2025
Upvotes: 141
Citations: 58

Read summary Original

16

arXiv.org

Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference

Seed Diffusion Preview is a large-scale discrete-state diffusion language model developed by ByteDance Seed and Tsinghua AIR, focused on code generation. It achieves an inference speed of 2,146 tokens per second on H20 GPUs, significantly faster than contemporary models like Mercury Coder and Gemini Diffusion, while maintaining competitive performance on…

Yuxuan Song, Zheng Zhang, Cheng Luo, Pengyang Gao, et al.

Published: Aug 2025
Upvotes: 139
Citations: 163

Read summary Original

17

arXiv.org

R-Zero: Self-Evolving Reasoning LLM from Zero Data

R-Zero is a fully autonomous framework for training reasoning LLMs without any pre-existing tasks or labels. It initializes two models from a single base LLM: a Challenger, trained via GRPO to generate questions at the edge of the Solver's capability (using an uncertainty reward based on self-consistency, plus repetition and format penalties), and a…

Chengsong Huang, Wenhao Yu, Xiaoyang Wang, Hongming Zhang, et al.

Published: Aug 2025
Upvotes: 134
Citations: 167

Read summary Original

18

arXiv.org

Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL

The paper introduces Chain-of-Agents (CoA), a paradigm enabling a single LLM to perform multi-agent problem-solving end-to-end by dynamically activating role-playing and tool agents. To train these Agent Foundation Models (AFMs), the authors propose multi-agent distillation, which converts trajectories from state-of-the-art multi-agent systems (e.g.,…

Weizhen Li, Jianbo Lin, Zhuosong Jiang, Jingyi Cao, et al.

Published: Aug 2025
Upvotes: 129
Citations: 66

Read summary Original

19

arXiv.org

rStar2-Agent: Agentic Reasoning Technical Report

rStar2-Agent is a 14B math reasoning model trained with agentic reinforcement learning (RL) to achieve frontier-level performance, rivaling or surpassing DeepSeek-R1 (671B). It uses Python coding tools within a dedicated environment, enabling advanced cognitive behaviors like careful tool use and reflection on execution feedback. Key innovations include an…

Ning Shang, Yifei Liu, Yi Zhu, Li Lyna Zhang, et al.

Published: Aug 2025
Upvotes: 120
Citations: 45

Read summary Original

20

arXiv.org

Beyond Pass@1: Self-Play with Variational Problem Synthesis Sustains RLVR

The paper proposes an online Self-play with Variational problem Synthesis (SvS) strategy for Reinforcement Learning with Verifiable Rewards (RLVR) to address entropy collapse and plateaued Pass@k performance in LLM reasoning. Standard RLVR improves Pass@1 but reduces generation diversity, limiting Pass@k. SvS uses the policy's correct solutions to…

Xiao Liang, Zhongzhi Li, Yeyun Gong, Yelong Shen, et al.

Published: Aug 2025
Upvotes: 119
Citations: 53

Read summary Original

21

Annual Meeting of the Association for Computational Linguistics

ReasonRank: Empowering Passage Ranking with Strong Reasoning Ability

ReasonRank is a reasoning-intensive listwise passage reranker that outperforms existing baselines on reasoning-heavy IR benchmarks. The authors address the scarcity of reasoning-intensive training data by proposing an automated synthesis framework that collects queries from diverse domains (complex QA, coding, math, web search) and uses DeepSeek-R1 to…

Wenhan Liu, Xinyu Ma, Weiwei Sun, Yutao Zhu, et al.

Published: Aug 2025
Upvotes: 117
Citations: 47

Read summary Original

22

arXiv.org

Ovis2.5 Technical Report

Ovis2.5 is a multimodal large language model from Alibaba Group, succeeding Ovis2. It introduces native-resolution visual perception using a NaViT vision transformer, which processes images at their original resolutions to preserve fine details and global layout, crucial for complex charts. The model also adds an optional 'thinking mode' for deep…

Shiyin Lu, Yang Li, Yu Xia, Yuwei Hu, et al.

Published: Aug 2025
Upvotes: 116
Citations: 81

Read summary Original

23

arXiv.org

WideSearch: Benchmarking Agentic Broad Info-Seeking

The paper introduces WideSearch, a benchmark for evaluating LLM-based search agents on large-scale, broad information-seeking tasks. It comprises 200 manually curated questions (100 English, 100 Chinese) across 18 domains, requiring agents to collect and structure extensive atomic information into tables. A five-stage quality control pipeline ensures task…

Ryan Wong, Jiawei Wang, Junjie Zhao, Li Chen, et al.

Published: Aug 2025
Upvotes: 113
Citations: 53

Read summary Original

24

arXiv.org

R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning

The paper introduces R-4B, a multimodal large language model (MLLM) with auto-thinking capability, which adaptively decides whether to engage in step-by-step reasoning based on problem complexity. R-4B is trained in two stages: bi-mode annealing, which uses a curated dataset of reasoning and non-reasoning examples to equip the model with both thinking and…

Qi Yang, Bolin Ni, Shiming Xiang, Han Hu, et al.

Published: Aug 2025
Upvotes: 111
Citations: 16

Read summary Original

25

arXiv.org

A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems

This survey reviews the emerging paradigm of self-evolving AI agents, which aim to bridge static foundation models with lifelong adaptability. It introduces a unified framework with four components: system inputs, agent system, environment, and optimizers, and proposes 'Three Laws' (Endure, Excel, Evolve) to guide safe and effective evolution. The survey…

Jinyuan Fang, Yanwen Peng, Xi Zhang, Yingxu Wang, et al.

Published: Aug 2025
Upvotes: 100
Citations: 154

Read summary Original

26

arXiv.org

SSRL: Self-Search Reinforcement Learning

The paper introduces Self-Search RL (SSRL), a method that trains LLMs to answer search-based questions using only their internal knowledge, without external search engines. The authors first quantify LLMs' intrinsic search capability via repeated sampling (Self-Search), showing strong scaling with inference budget and high pass@k on benchmarks like…

Yuchen Fan, Kaiyan Zhang, Heng Zhou, Yuxin Zuo, et al.

Published: Aug 2025
Upvotes: 97
Citations: 11

Read summary Original

27

Meta AI

Deep Think with Confidence

Deep Think with Confidence (DeepConf) is a test-time method that improves LLM reasoning efficiency and accuracy by using model-internal confidence signals to filter low-quality reasoning traces. It addresses the diminishing returns and high computational overhead of self-consistency with majority voting. DeepConf operates in offline mode (filtering…

Yichao Fu, Xuewei Wang, Yuandong Tian, Jiawei Zhao

Published: Aug 2025
Upvotes: 92
Citations: 151

Read summary Original

28

AAAI Conference on Artificial Intelligence

Beyond Transcription: Mechanistic Interpretability in ASR

This paper adapts interpretability methods from LLMs—logit lens, linear probing, and activation patching—to analyze the internal mechanisms of ASR models, specifically Whisper-large-v3 and Qwen2-Audio. The authors find that acoustic and semantic attributes (e.g., speaker gender, noise, accent) are linearly decodable from encoder layers, with peak…

Neta Glazer, Yael Segal-Feldman, Hilit Segev, Aviv Shamsian, et al.

Published: Aug 2025
Upvotes: 92
Citations: 11

Read summary Original

29

arXiv.org

Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning

The paper introduces Pref-GRPO, a reinforcement learning method for text-to-image (T2I) generation that replaces pointwise reward score maximization with pairwise preference fitting to address reward hacking. It identifies the root cause of reward hacking as an 'illusory advantage' arising from normalizing tightly clustered pointwise scores by a small…

Yibin Wang, Zhimin Li, Yuhang Zang, Yujie Zhou, et al.

Published: Aug 2025
Upvotes: 91
Citations: 90

Read summary Original

30

arXiv.org

DuPO: Enabling Reliable LLM Self-Verification via Dual Preference Optimization

DuPO is a dual learning-based preference optimization framework that generates annotation-free feedback for LLM optimization. It addresses limitations of RLVR (costly labels, restricted to verifiable tasks) and traditional dual learning (strictly dual task pairs). DuPO decomposes a primal task's input into known and unknown components, then constructs a…

Shuaijie She, Yu Bao, Yu Lu, Lu Xu, et al.

Published: Aug 2025
Upvotes: 86
Citations: 5

Read summary Original

31

arXiv.org

Self-Rewarding Vision-Language Model via Reasoning Decomposition

Vision-SR1 is a three-stage self-rewarding reinforcement learning method for vision-language models (VLMs) that improves visual reasoning without external visual supervision. It decomposes VLM reasoning into visual perception and language reasoning, prompting the model to first generate a self-contained visual description, then re-prompting it to answer…

Zongxia Li, Wenhao Yu, Chengsong Huang, Zhenwen Liang, et al.

Published: Aug 2025
Upvotes: 85
Citations: 91

Read summary Original

32

arXiv.org

LLaVA-Critic-R1: Your Critic Model is Secretly a Strong Policy Model

The paper introduces LLaVA-Critic-R1, a multimodal critic model trained via reinforcement learning (RL) on preference-labeled critic data, which surprisingly also becomes a strong policy model. By reformulating critic data into verifiable tasks and using GRPO with preference and format rewards, the model improves over its base (Qwen-2.5-VL-7B) by +5.7% on…

Xiyao Wang, Chunyuan Li, Jianwei Yang, Kai Zhang, et al.

Published: Aug 2025
Upvotes: 85
Citations: 30

Read summary Original

33

arXiv.org

Thyme: Think Beyond Images

The paper introduces Thyme (Think Beyond Images), a paradigm for multimodal large language models (MLLMs) that autonomously generates and executes code for image manipulations (cropping, rotation, contrast enhancement) and complex computations, going beyond existing 'think with images' approaches. Thyme uses a two-stage training: Supervised Fine-Tuning…

Yi-Fan Zhang, Xingyu Lu, Shukang Yin, Chaoyou Fu, et al.

Published: Aug 2025
Upvotes: 81
Citations: 119

Read summary Original

34

arXiv.org

TreePO: Bridging the Gap of Policy Optimization and Efficacy and Inference Efficiency with Heuristic Tree-based Modeling

TreePO is a reinforcement learning framework for large language models that replaces standard independent trajectory sampling with a heuristic tree-based rollout mechanism. It models sequence generation as a tree search with dynamic branching and fixed-length segment decoding, leveraging shared prefixes via KV-cache reuse to reduce compute. The method…

Yizhi Li, Qingshui Gu, Zhoufutu Wen, Ziniu Li, et al.

Published: Aug 2025
Upvotes: 80
Citations: 59

Read summary Original

35

Research paper

EO-1: An Open Unified Embodied Foundation Model for General Robot Control

The paper introduces EO-Robotics, comprising the EO-1 model and the EO-Data1.5M dataset, to advance general robot control. EO-1 is a 3B-parameter unified embodied foundation model that integrates multimodal understanding and robot action generation in a single decoder-only transformer, combining autoregressive decoding for text with flow-matching denoising…

Delin Qu, Haoming Song, Qizhi Chen, Zhaoqing Chen, et al.

Published: Aug 2025
Upvotes: 78
Citations: 21

Read summary Original

36

arXiv.org

Matrix-3D: Omnidirectional Explorable 3D World Generation

Matrix-3D is a framework for generating omnidirectional, explorable 3D worlds from a single image or text prompt. It uses panoramic representations to overcome the limited field of view of perspective-based methods. The pipeline first generates a panorama image, then a trajectory-guided panoramic video using a diffusion model conditioned on scene mesh…

Zhongqi Yang, Wenhang Ge, Yuqi Li, Jiaqi Chen, et al.

Published: Aug 2025
Upvotes: 76
Citations: 28

Read summary Original

37

AAAI Conference on Artificial Intelligence

ComoRAG: A Cognitive-Inspired Memory-Organized RAG for Stateful Long Narrative Reasoning

ComoRAG is a cognitive-inspired, memory-organized RAG framework for stateful long narrative reasoning. It addresses the limitations of traditional RAG methods, which are stateless and single-step, by mimicking the human Prefrontal Cortex's Metacognitive Regulation. The framework uses a dynamic memory workspace and an iterative cognitive loop with five…

Juyuan Wang, Rongchen Zhao, Wei Wei, Yufeng Wang, et al.

Published: Aug 2025
Upvotes: 75
Citations: 13

Read summary Original

38

arXiv.org

Droplet3D: Commonsense Priors from Videos Facilitate 3D Generation

Droplet3D addresses 3D data scarcity by leveraging commonsense priors from videos for 3D generation. The authors introduce Droplet3D-4M, a large-scale dataset of 4 million 3D models, each with an 85-frame 360-degree orbital rendering video and dense multi-view-level text captions averaging 260 words. They also present Droplet3D, a generative model…

Xiaochuan Li, Guoguang Du, Runze Zhang, Liang Jin, et al.

Published: Aug 2025
Upvotes: 75
Citations: 2

Read summary Original

39

arXiv.org

Open Data Synthesis For Deep Research

This technical report introduces InfoSeek, a scalable framework for synthesizing Deep Research tasks, formalized as Hierarchical Constraint Satisfaction Problems (HCSPs). Unlike simpler multi-hop or flat CSP problems, HCSPs require traversing a hierarchy of interdependent constraints. InfoSeek uses a dual-agent system (Planner and Browser) to recursively…

Ziyi Xia, Kun Luo, Hongjin Qian, Zheng Liu

Published: Aug 2025
Upvotes: 74
Citations: 14

Read summary Original

40

arXiv.org

Genie Envisioner: A Unified World Foundation Platform for Robotic Manipulation

Genie Envisioner (GE) is a unified world foundation platform for robotic manipulation, integrating policy learning, evaluation, and simulation within a video-generative framework. GE-Base, an instruction-conditioned multi-view video diffusion model, is pretrained on ~3,000 hours of data from one million real-world episodes (AgiBot-World-Beta) to capture…

Yue Liao, Pengfei Zhou, Siyuan Huang, Donglin Yang, et al.

Published: Aug 2025
Upvotes: 73
Citations: 113

Read summary Original

41

arXiv.org

FutureX: An Advanced Live Benchmark for LLM Agents in Future Prediction

FutureX is a live benchmark for evaluating LLM agents on future prediction tasks, built to avoid data contamination by using only future events. It collects questions daily from 195 curated websites across 11 domains, runs 25 models (base LLMs, Think&Search models, open-source and closed-source Deep Research agents), and automatically scores predictions…

Zhiyuan Zeng, Jiashuo Liu, Siyuan Chen, Tianci He, et al.

Published: Aug 2025
Upvotes: 73
Citations: 39

Read summary Original

42

arXiv.org

PRELUDE: A Benchmark Designed to Require Global Comprehension and Reasoning over Long Contexts

PRELUDE is a benchmark for evaluating long-context understanding in LLMs by asking whether a character's prequel story is consistent with the canonical narrative of a book. The task requires global comprehension and deep reasoning, as 88% of instances need evidence from multiple parts of the narrative. The dataset includes 795 instances across 13 books,…

Mo Yu, Tsz Ting Chung, Chulun Zhou, Tong Li, et al.

Published: Aug 2025
Upvotes: 71
Citations: 6

Read summary Original

43

arXiv.org

Story2Board: A Training-Free Approach for Expressive Storyboard Generation

Story2Board is a training-free framework for generating expressive storyboards from natural language. It introduces two mechanisms: Latent Panel Anchoring (LPA), which pairs each panel with a shared reference prompt to maintain character consistency, and Reciprocal Attention Value Mixing (RAVM), which blends value vectors between semantically aligned…

David Dinkevich, Matan Levy, Omri Avrahami, Dvir Samuel, et al.

Published: Aug 2025
Upvotes: 70
Citations: 12

Read summary Original

44

arXiv.org

MeshCoder: LLM-Powered Structured Mesh Code Generation from Point Clouds

MeshCoder is a framework that reconstructs 3D objects from point clouds into editable Blender Python scripts. It introduces a set of expressive Blender Python APIs capable of modeling complex geometries beyond simple primitives, including primitives, translation, bridge loops, boolean operations, and arrays. A large-scale paired object-code dataset was…

Bingquan Dai, Li Ray Luo, Qihong Tang, Jie Wang, et al.

Published: Aug 2025
Upvotes: 69
Citations: 12

Read summary Original

45

arXiv.org

AgentScope 1.0: A Developer-Centric Framework for Building Agentic Applications

AgentScope 1.0 is a developer-centric framework for building agentic applications, grounded in the ReAct paradigm. It provides foundational components (message, model, memory, tool) with unified interfaces and extensible modules, supporting multimodal messages, diverse LLM providers, and MCP integration. The framework includes agent-level infrastructure…

Dawei Gao, Zitao Li, Yuexiang Xie, Weirui Kuang, et al.

Published: Aug 2025
Upvotes: 68
Citations: 11

Read summary Original

46

arXiv.org

Mobile-Agent-v3: Fundamental Agents for GUI Automation

The paper introduces GUI-Owl, a foundational GUI agent model built on Qwen2.5-VL, and Mobile-Agent-v3, a multi-agent framework. GUI-Owl-7B achieves state-of-the-art scores among open-source models, including 66.4 on AndroidWorld and 34.9 on OSWorld-Verified, while Mobile-Agent-v3 further improves these to 73.3 and 37.7, respectively. Key innovations…

Jiabo Ye, Xi Zhang, Haiyang Xu, Haowei Liu, et al.

Published: Aug 2025
Upvotes: 66
Citations: 150

Read summary Original

47

arXiv.org

Skywork UniPic: Unified Autoregressive Modeling for Visual Understanding and Generation

Skywork UniPic is a 1.5B-parameter autoregressive model that unifies image understanding, text-to-image generation, and image editing in a single architecture without task-specific adapters. It achieves a GenEval score of 0.86, a DPG-Bench score of 85.5, and GEditBench-EN and ImgEdit-Bench scores of 5.83 and 3.49, respectively, while generating 1024x1024…

Peiyu Wang, Yi Peng, Yimeng Gan, Liang Hu, et al.

Published: Aug 2025
Upvotes: 65
Citations: 17

Read summary Original

48

AAAI Conference on Artificial Intelligence

DeepPHY: Benchmarking Agentic VLMs on Physical Reasoning

DeepPHY is a benchmark framework for evaluating interactive physical reasoning in agentic Vision-Language Models (VLMs). It integrates six physics-based simulation environments: PHYRE, I-PHYRE, Kinetix, Pooltool, Angry Birds, and Cut the Rope, converting continuous action spaces into discrete, structured formats and augmenting observations with…

Xinrun Xu, Pi Bu, Ye Wang, Börje F. Karlsson, et al.

Published: Aug 2025
Upvotes: 65
Citations: 10

Read summary Original

49

AAAI Conference on Artificial Intelligence

Omni-Effects: Unified and Spatially-Controllable Visual Effects Generation

Omni-Effects is a unified framework for generating spatially controllable visual effects (VFX) in videos, addressing limitations of existing per-effect LoRA training. It introduces two key innovations: LoRA-based Mixture of Experts (LoRA-MoE) to integrate diverse effects in a single model while mitigating cross-task interference, and Spatial-Aware Prompt…

Fangyuan Mao, Aiming Hao, Jintao Chen, Dongxia Liu, et al.

Published: Aug 2025
Upvotes: 63
Citations: 28

Read summary Original

50

arXiv.org

MCP-Bench: Benchmarking Tool-Using LLM Agents with Complex Real-World Tasks via MCP Servers

MCP-Bench is a benchmark for evaluating LLM agents on realistic, multi-step tool-use tasks via the Model Context Protocol (MCP). It connects agents to 28 live MCP servers with 250 tools across domains like finance, travel, and science, enabling complex intra- and cross-server workflows. Tasks are generated via an LLM-based synthesis pipeline that discovers…

Zhenting Wang, Qi Chang, Hemani Patel, Shashank Biju, et al.

Published: Aug 2025
Upvotes: 63
Citations: 90

Read summary Original

51

arXiv.org

4DNeX: Feed-Forward 4D Generative Modeling Made Easy

4DNeX is the first feed-forward framework for generating 4D (dynamic 3D) scene representations from a single image. It fine-tunes a pretrained video diffusion model (Wan2.1) to generate a unified 6D video representation (RGB and XYZ sequences) jointly, enabling efficient end-to-end image-to-4D generation. To address data scarcity, the authors constructed…

Zhaoxi Chen, Tianqi Liu, Long Zhuo, Jiawei Ren, et al.

Published: Aug 2025
Upvotes: 62
Citations: 31

Read summary Original

52

ACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia

Voost: A Unified and Scalable Diffusion Transformer for Bidirectional Virtual Try-On and Try-Off

Voost is a unified and scalable diffusion transformer framework that jointly learns virtual try-on and try-off using a single model. It uses a token-level concatenation of garment and person images, with a task token encoding generation direction and garment category, enabling bidirectional supervision without task-specific networks or auxiliary losses.…

Seungyong Lee, Jeong-gi Kwak

Published: Aug 2025
Upvotes: 60
Citations: 12

Read summary Original

53

arXiv.org

BeyondWeb: Lessons from Scaling Synthetic Data for Trillion-scale Pretraining

BeyondWeb is a synthetic data generation framework for LLM pretraining that outperforms existing synthetic datasets like Cosmopedia and Nemotron-Synth by up to 5.1 and 2.6 percentage points, respectively, across 14 benchmarks. It achieves up to 7.7x faster training than open web data and 2.7x faster than Nemotron-Synth. A 3B model trained on BeyondWeb for…

DatologyAI, :, Pratyush Maini, Vineeth Dorna, et al.

Published: Aug 2025
Upvotes: 60
Citations: 16

Read summary Original

54

arXiv.org

Training Long-Context, Multi-Turn Software Engineering Agents with Reinforcement Learning

The paper presents a two-phase training pipeline for software engineering (SWE) agents using reinforcement learning (RL). Starting from Qwen2.5-72B-Instruct, the authors first apply rejection fine-tuning (RFT) on successful trajectories from SWE-rebench tasks, improving Pass@1 on SWE-bench Verified from 11% to 20%. Then, they apply a synchronous RL…

Alexander Golubev, Maria Trofimova, Sergei Polezhaev, Ibragim Badertdinov, et al.

Published: Aug 2025
Upvotes: 59
Citations: 27

Read summary Original

55

arXiv.org

From Scores to Skills: A Cognitive Diagnosis Framework for Evaluating Financial Large Language Models

The paper introduces FinCDM, the first cognitive diagnosis evaluation framework for financial large language models (LLMs), moving beyond aggregate scores to assess knowledge-skill level proficiency. It constructs CPA-KQA, a dataset of 210 expert-annotated questions derived from the CPA exam, covering 70 financial concepts, with high inter-annotator…

Ziyan Kuang, Feiyu Zhu, Maowei Jiang, Yanzhao Lai, et al.

Published: Aug 2025
Upvotes: 59
Citations: 4

Read summary Original

56

IEEE International Conference on Computer Vision

LongSplat: Robust Unposed 3D Gaussian Splatting for Casual Long Videos

LongSplat is a framework for novel view synthesis from casually captured long videos without known camera poses. It jointly optimizes camera poses and 3D Gaussian Splatting (3DGS) to address pose drift, inaccurate geometry initialization, and memory limitations. Key components include incremental joint optimization, a pose estimation module using learned…

Chin-Yang Lin, Cheng Sun, Fu-En Yang, Min-Hung Chen, et al.

Published: Aug 2025
Upvotes: 59
Citations: 28

Read summary Original

57

arXiv.org

Seeing, Listening, Remembering, and Reasoning: A Multimodal Agent with Long-Term Memory

The paper introduces M3-Agent, a multimodal agent framework with long-term memory that processes real-time video and audio to build episodic and semantic memories, organized in an entity-centric multimodal graph. It uses reinforcement learning for multi-turn reasoning and iterative memory retrieval. The authors also present M3-Bench, a long-video question…

Lin Long, Yichen He, Wentao Ye, Yiyuan Pan, et al.

Published: Aug 2025
Upvotes: 58
Citations: 64

Read summary Original

58

arXiv.org

Hermes 4 Technical Report

Hermes 4 is a family of hybrid reasoning models (14B, 70B, 405B) that combine structured, multi-turn reasoning with broad instruction-following. The models were trained on a dataset of ~5 million samples (19B tokens), including 3.5M reasoning and 1.6M non-reasoning samples, synthesized via DataForge, a graph-based generator, and rejection sampling with…

Ryan Teknium, Roger Jin, Jai Suphavadeeprasit, Dakota Mahan, et al.

Published: Aug 2025
Upvotes: 56
Citations: 3

Read summary Original

59

arXiv.org

USO: Unified Style and Subject-Driven Generation via Disentangled and Reward Learning

The paper introduces USO, a unified framework for style-driven and subject-driven image generation, arguing these tasks are complementary and can be jointly optimized through cross-task co-disentanglement. The authors construct a large-scale triplet dataset (content, style, stylized images) using a subject-for-style curation framework with stylization and…

Shaojin Wu, Mengqi Huang, Yufeng Cheng, Wenxu Wu, et al.

Published: Aug 2025
Upvotes: 56
Citations: 29

Read summary Original

60

arXiv.org

ELV-Halluc: Benchmarking Semantic Aggregation Hallucinations in Long Video Understanding

The paper introduces ELV-Halluc, the first benchmark for evaluating Semantic Aggregation Hallucination (SAH) in long videos. SAH occurs when models correctly perceive frame-level semantics but misattribute them across events, a problem that intensifies with semantic complexity. The benchmark uses event-by-event videos (average 672.4 seconds) and…

Hao Lu, Jiahao Wang, Yaolun Zhang, Ruohui Wang, et al.

Published: Aug 2025
Upvotes: 55
Citations: 10

Read summary Original

61

arXiv.org

ToonComposer: Streamlining Cartoon Production with Generative Post-Keyframing

ToonComposer is a generative model that unifies the traditionally separate inbetweening and colorization stages of cartoon production into a single post-keyframing stage. Built on the DiT-based video foundation model Wan 2.1, it uses a sparse sketch injection mechanism for precise control from keyframe sketches and a spatial low-rank adapter (SLRA) to…

Lingen Li, Guangzhi Wang, Zhaoyang Zhang, Yaowei Li, et al.

Published: Aug 2025
Upvotes: 54
Citations: 8

Read summary Original

62

arXiv.org

LongVie: Multimodal-Guided Controllable Ultra-Long Video Generation

LongVie is a framework for controllable ultra-long video generation, addressing temporal inconsistency and visual degradation in autoregressive generation. It identifies three key issues: separate noise initialization, independent control signal normalization, and single-modality guidance limitations. LongVie introduces unified noise initialization and…

Jianxiong Gao, Zhaoxi Chen, Xian Liu, Jianfeng Feng, et al.

Published: Aug 2025
Upvotes: 53
Citations: 19

Read summary Original

63

arXiv.org

Speed Always Wins: A Survey on Efficient Architectures for Large Language Models

This survey systematically reviews efficient architectures for Large Language Models (LLMs), addressing the computational bottlenecks of the standard Transformer, particularly the quadratic complexity of self-attention and the high costs of large Feed-Forward Networks. It categorizes recent innovations into seven main areas: linear sequence modeling (e.g.,…

Weigao Sun, Jiaxi Hu, Yucheng Zhou, Jusen Du, et al.

Published: Aug 2025
Upvotes: 53
Citations: 17

Read summary Original